Acme-Lou

 view release on metacpan or  search on metacpan

author/en2kana.csv  view on Meta::CPAN

maker,メーカー
making,メイキング
mama,ママ
man,マン
manage,マネージ
manager,マネージャー
mankind,マンカインド
manner,マナー
manual,マニュアル
many,メニー
map,マップ
march,マーチ
mark,マーク
market,マーケット
married,マリード
marry,マリー
marvelous,マーベラス
mass,マス
master,マスター
match,マッチ
mate,メイト

author/ja2kana.pl  view on Meta::CPAN

my %en2kana;
open my $en2kana, '<:encoding(utf8)', "$Bin/lou-en2kana.csv";
while (<$en2kana>) {
    chomp;
    next unless $_;
    next if /^#/;
    my ($en, $kana) = split ',';
    $en2kana{lc $en} = $kana;
}

my %skip_word = map { $_ => 1 } qw(
    now say new be come see is as
    one two three four five
    six seven eight nine ten
    law raw row whole weigh
    hurt hut firm fare flesh
    youth lack role waste worth
    few pray health sex
);

warn "make $Bin/lou-ja2kana.csv...\n";

author/loucsv.pl  view on Meta::CPAN

useage:
    cat mecab-ipadic-2.7.0-20060707/*.csv | perl loucsv.pl > lou.csv

original:
    http://chasen.org/~taku/blog/archives/2007/01/_mecab.html

=cut

my $ja2kana = get_ja2kana();

my $pos_form_okuri_map = {
    '名詞-*'                => '',
    '感動詞-*'              => '',
    '接続詞-*'              => '',
    '連体詞-*'              => '',
    '動詞-仮定形'           => 'すれ',
    '動詞-仮定縮約1'       => 'すれ',
    '動詞-基本形'           => 'する',
    '動詞-体言接続'         => 'する',
    '動詞-体言接続特殊2'   => 'す',
    '動詞-文語基本形'       => 'する',

author/loucsv.pl  view on Meta::CPAN


    # 表層形,左文脈ID,右文脈ID,コスト,品詞,品詞細分類1,品詞細分類2,品詞細分類3,活用型,活用形,原形,読み,発音
    my ($c, $d, $e, $cost, $pos, $f, $g, $h, $type, $form, $lemma, $i, $j) = split /,/, $feature;
    
    my $kana = $ja2kana->{$lemma};
    if (not defined $kana) {
        return;
    }
    
    my $pos_form = "$pos-$form";
    my $okuri = $pos_form_okuri_map->{$pos_form}; 
    if (not defined $okuri) {
        return;
    }

    $cost -= 1000;
    $cost = 1 if $cost <= 0;
    
    return join ",", $c, $d, $e, $cost, $pos, $f, $g, $h, $type, $form, $lemma, $i, $j, $kana, $okuri, "\n";
}



( run in 1.015 second using v1.01-cache-2.11-cpan-49f99fa48dc )